Intelligent machining system of numerical control machine tool automatic production line based on cloud computing

Abstract

The invention discloses an intelligent processing system of a numerical control machine tool automatic production line based on cloud computing, and relates to the technical field of numerical control machine tool automatic production, the system comprises the following components: a data acquisition layer, a cloud computing processing layer, a task planning layer and an intelligent scheduling execution layer; according to the method, a biological heuristic intelligent scheduling algorithm is adopted, a pheromone exchange mechanism simulating an ant colony algorithm and a division cooperation mode simulating a bee colony are fused, an accurate scheduling scheme can be generated, the algorithm can adaptively adjust a scheduling strategy according to the dynamic change and task characteristics of a production environment, and the scheduling efficiency is improved. Meanwhile, the system further has a machine tool fault prediction function, potential faults can be found in advance, early warning information can be sent, production tasks can be adjusted in time, production interruption caused by the machine tool faults is avoided, the functions jointly act on the production line, the flexibility and stability of the system are greatly improved, and the production efficiency is improved. And the production cost and risk are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field, and in particular to an intelligent processing system for an automated production line of CNC machine tools based on cloud computing. Background Art

[0002] With the rapid development of the manufacturing industry and increasingly fierce market competition, CNC machine tool automation production lines play a vital role in industrial production. However, traditional CNC machine tool automation production lines have many shortcomings in data processing, task planning and scheduling execution, and it is difficult to meet the modern manufacturing industry's needs for efficiency, flexibility and intelligence.

[0003] Traditional technology has its shortcomings. On the one hand, traditional technology cannot generate accurate scheduling plans and cannot avoid bottlenecks and waste in the production process. On the other hand, traditional technology cannot detect potential faults in advance and send early warning information, and cannot adjust production tasks in time to avoid production interruptions caused by machine tool failures. Therefore, it is particularly important to develop an intelligent processing system for CNC machine tool automation production lines based on cloud computing. Summary of the invention

[0004] The purpose of the present invention is to make up for the shortcomings of the prior art and to provide an intelligent processing system for an automated production line of CNC machine tools based on cloud computing. It can adaptively adjust strategies to cope with changes in the production environment and task characteristics, avoid bottlenecks and waste, and has the function of predicting machine tool faults, giving early warning of potential faults, enhancing system flexibility and stability, and reducing production costs and risks by adopting a bio-inspired intelligent scheduling algorithm that integrates pheromone communication of simulated ant colony algorithms and the division of labor and cooperation mode of bee colonies. It also has the function of predicting machine tool faults, giving early warning of potential faults, enhancing system flexibility and stability, and reducing production costs and risks.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: an intelligent processing system for a CNC machine tool automated production line based on cloud computing, the system comprising the following components: a data acquisition layer, a cloud computing processing layer, a task planning layer, and an intelligent scheduling execution layer; The data acquisition layer is deployed at each key node of the CNC machine tool automation production line and is composed of a variety of sensors and data acquisition equipment. This layer is responsible for collecting all-round and real-time production-related data covering CNC machine tool operation data, production task information, and multi-dimensional raw material status. For CNC machine tool operation data, the collected parameters include but are not limited to the spindle speed. , Feed speed , tool wear , Machine tool load power , by installing corresponding sensors at key parts of the machine tool, obtaining production task information, and recording the task number in detail , Task Type , Processing requirements , Estimated processing time , Task Priority , raw material status data involves the type of raw materials , Inventory quantity , quality inspection parameters ,The collected data is transmitted in real time to the cloud computing processing layer in encrypted form through a high-speed and stable network transmission protocol; The cloud computing processing layer: relying on a powerful cloud computing platform, has large-scale data storage and high-performance computing capabilities. After receiving the data transmitted by the data acquisition layer, it first performs data cleaning operations on it, removes noise data and obviously erroneous data records by setting data threshold ranges and abnormal data recognition algorithms, and then classifies and organizes them according to data types and features, and stores the processed data in a distributed database. For different types of data, it uses specific data analysis algorithms for in-depth mining. For machine tool operation data, it establishes a machine tool performance evaluation model ,in , , , is the weight coefficient. According to the historical operation data of the machine tool and the actual processing experience, the hierarchical analysis method is used to determine the weight coefficients to evaluate the current performance status of the machine tool. For the production task information, the task complexity evaluation formula is used , As the weight coefficient, various types of information obtained from the analysis are integrated to provide comprehensive and accurate data support for the task planning layer; The task planning layer: Based on the detailed data support provided by the cloud computing processing layer, it conducts comprehensive analysis and planning of production tasks, and , Estimated processing time As well as the logical relationship between tasks, build a task priority sorting model ,in The comprehensive priority ranking result of the tasks is is the number of factors that affect task priority, is the weight coefficient of each factor, For the Factors that affect task priority. For urgent order tasks, priority factors are given higher weights. According to the performance evaluation results of the machine tool Based on the availability of resources, a preliminary task allocation plan is formulated to reasonably allocate tasks to different processing stages and machine tool groups, providing a task planning framework for the intelligent scheduling execution layer; The intelligent scheduling execution layer: based on the task allocation framework provided by the task planning layer and the analysis results of the cloud computing processing layer, uses the biologically inspired intelligent scheduling algorithm to generate an accurate scheduling plan. The algorithm combines the pheromone communication mechanism of the simulated ant colony algorithm and the division of labor and cooperation model of the bee colony. In terms of simulating the pheromone communication mechanism of the ant colony algorithm, the pheromone concentration update formula is defined: ,in express Time Machine Tool To the task The pheromone concentration, is the pheromone volatility coefficient, with a value range of (0, 1). Its optimal value is determined through experiments and data analysis according to the dynamic changes of the production environment and the characteristics of the task. for Time Machine Tool Complete the task The amount of pheromone released after , is the total amount of pheromone released, For machine tools Complete the task The path length is used to simulate the division of labor and cooperation mode of the bee colony to build a resource allocation model: ,in Representation Resources Assign to task The adaptability, For resources The own attribute parameters, For the task The demand parameters, For resources With the task The matching parameters, , , is the weight coefficient. The intelligent scheduling execution layer sends detailed scheduling instructions to each CNC machine tool for execution based on the results generated by the above algorithm.

[0006] Furthermore, the data acquisition device of the data acquisition layer uses different acquisition frequencies for different types of parameters when collecting the operation data of the CNC machine tool. and feed speed For parameters with high real-time requirements and frequent changes, set the acquisition frequency to times / second, accurately collects data through high-speed data acquisition cards, and measures tool wear and machine load power For parameters that change relatively slowly, the acquisition frequency is set to The data is collected times / minute using a timed sampling method. To ensure the accuracy and reliability of the data, the collected data is verified and corrected in real time. The data is encoded and decoded using a redundant check algorithm and an error detection code. When a data error is detected, re-sampling or repair operations are automatically performed. The data acquisition equipment also has an adaptive acquisition function that can automatically adjust the acquisition frequency and accuracy according to the operating status of the machine tool and the characteristics of the production task, so as to meet the data acquisition needs in different production scenarios.

[0007] Furthermore, when performing data cleaning, the cloud computing processing layer adopts different cleaning strategies for different types of data. For numerical data, by setting upper and lower threshold ranges, outliers that exceed the range are removed. For character data, regular expression matching and semantic analysis methods are used to identify and correct spelling errors and data in irregular formats. In the process of data classification and organization, the data is divided into multiple different categories according to its source, nature and purpose, and an independent data table structure is established for each category to facilitate subsequent data query and analysis. In addition, the cloud computing processing layer also has data backup and recovery functions, regularly backs up the processed data in full, and records data change operations in real time. When data is lost or damaged, it can quickly and accurately restore the data to ensure data integrity and availability.

[0008] Furthermore, when the cloud computing processing layer uses the data analysis algorithm to analyze the machine tool operation data, in addition to establishing the machine tool performance evaluation model In addition, a machine tool fault prediction model was built. This model is based on a machine learning algorithm. By training a large amount of historical machine tool operation data and fault records, characteristic parameters related to machine tool faults are extracted, and a support vector machine algorithm is used to establish a fault prediction model. ,in is the extracted feature parameter, The mapping function obtained through training is used to improve the accuracy and reliability of fault prediction by continuously optimizing model parameters and adjusting the combination of feature parameters. When a machine tool failure is predicted, early warning information is sent to the intelligent scheduling execution layer in a timely manner to adjust production tasks in advance and avoid production interruptions caused by machine tool failures.

[0009] Furthermore, the task planning layer constructs a task priority sorting model When considering factors that affect task priority Perform more detailed analysis, in addition to task priority , Estimated processing time In addition, we also consider the special requirements of customer orders and the urgency of market demand. We give higher priority to orders with special process requirements or tight delivery dates. At the same time, we combine market demand data to give priority to product tasks that are urgently needed in the market. At the same time, through market research, customer feedback and comprehensive analysis of historical order data, we continuously optimize the weight distribution to make the task priority sorting more in line with actual production needs and market orientation.

[0010] Furthermore, when formulating the preliminary task allocation plan, the task planning layer fully considers the maintenance plan and production capacity balance of the machine tools. For the machine tools that are about to undergo maintenance, the task volume is minimized when allocating tasks to avoid production interruptions due to maintenance. At the same time, by analyzing the performance parameters and historical processing data of the machine tools, the production capacity of different machine tools is quantitatively evaluated, and tasks are reasonably allocated to different machine tools to achieve production load balance among the machine tools. For high-precision and high-speed machine tools, tasks with high precision requirements and high processing difficulty are allocated. For ordinary machine tools, conventional processing tasks are allocated. In addition, the task planning layer will arrange the start time of the task according to the inventory situation and supply plan of the raw materials to ensure the timely supply of raw materials and avoid task delays due to shortage of raw materials.

[0011] Furthermore, when the intelligent scheduling execution layer simulates the pheromone communication mechanism of the ant colony algorithm, in order to prevent the algorithm from falling into a local optimal solution, a random disturbance factor is introduced. During the pheromone update process, the pheromone concentration is adjusted. Perform random adjustment, the adjustment formula is ,in is the random disturbance factor, and its value range is , is the disturbance intensity coefficient. According to the complexity of the production task and the scale of the production line, its optimal value is determined through experimental tests. At the same time, in order to speed up the convergence of the algorithm, the path with lower pheromone concentration is penalized to reduce its probability of being selected. The specific implementation method is as follows: when the pheromone concentration of a certain path is Below the set threshold When the probability of selecting Reduced to the original times, , in this way, the algorithm is guided to search for the global optimal solution faster.

[0012] Furthermore, the intelligent scheduling execution layer constructs a resource allocation model by imitating the division of labor and cooperation mode of the bee colony. When the resource's own attribute parameters Further refinement, for CNC machine tools, resource attribute parameters include not only basic parameters of accuracy and power, but also maintenance records and service life of the machine tools. Maintenance records are quantified by the number of maintenance times and maintenance time interval indicators, and service life is calculated based on the production date and current time of the machine tools. Incorporating these factors into resource attribute parameters can more comprehensively evaluate the actual processing capacity and reliability of the machine tools. , , In addition to using correlation analysis and optimization algorithms to solve the problem, expert experience weights are also introduced. Industry experts are organized to evaluate the importance of resources and tasks, assign different weights, and then merge them with the weights obtained from data analysis to obtain the final weight coefficient, so as to improve the accuracy and rationality of the resource allocation model.

[0013] Compared with the existing technology, the intelligent processing system of the CNC machine tool automated production line based on cloud computing has the following beneficial effects: 1. The system adopts a bio-inspired intelligent scheduling algorithm, which integrates the pheromone communication mechanism of the simulated ant colony algorithm and the division of labor and cooperation model of the bee colony. It can generate accurate scheduling plans. This algorithm can adaptively adjust the scheduling strategy according to the dynamic changes of the production environment and the characteristics of the task, thereby avoiding bottlenecks and waste in the production process. At the same time, the system also has the function of machine tool failure prediction, which can detect potential failures in advance and send early warning information, so as to adjust production tasks in time and avoid production interruptions caused by machine tool failures. These functions work together on the production line, greatly enhancing the flexibility and stability of the system and reducing production costs and risks.

[0014] 2. The system realizes efficient and intelligent management of CNC machine tool automation production lines through the close collaboration of data acquisition layer, cloud computing processing layer, task planning layer and intelligent scheduling execution layer. The system can collect production-related data in an all-round and real-time manner, and conduct in-depth mining and analysis of data through the large-scale data storage and high-performance computing capabilities of the cloud computing platform. This not only improves the accuracy and reliability of production data, but also provides comprehensive and accurate data support for the comprehensive analysis and planning of production tasks. Therefore, the system can significantly improve the operating efficiency and processing accuracy of the production line, bringing higher production benefits and product quality to the enterprise.

[0015] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 This is the process operation diagram of the intelligent processing system of the CNC machine tool automation production line based on cloud computing. DETAILED DESCRIPTION

[0018] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0019] Embodiment 1 This embodiment describes a factory that produces small precision parts with diverse product types, large order quantities and strict delivery requirements. The production line consists of multiple high-precision CNC machine tools, and an efficient intelligent processing system is required to ensure the accuracy and timeliness of production.

[0020] Install high-precision sensors on each CNC machine tool to monitor the spindle speed. and feed speed , using a high-speed data acquisition card to Real-time acquisition at a frequency of times per second ensures accurate capture of dynamic changes in the machining process and tool wear and machine load power Then The data is sampled and collected at a frequency of times per minute. The collected data is verified and corrected in real time through a redundant check algorithm and error detection code to ensure data accuracy. For example, when processing a batch of high-precision shaft parts, the data acquisition equipment accurately records the changes in spindle speed at different processing stages, as well as the growth of tool wear with processing time.

[0021] During data cleaning, for numerical data such as spindle speed and feed speed, a reasonable threshold range is set to remove abnormal values. For character data such as task type, regular expression matching and semantic analysis are used to correct errors. After classification and sorting, the machine tool operation data and production task information are stored in different data tables respectively, and the machine tool performance evaluation model is used. (Assumption , , , ) Evaluate machine tool performance. For example, during the machining process, a machine tool can calculate the value, judge its performance status, and build a machine tool fault prediction model to detect potential faults in advance.

[0022] Build a task prioritization model , taking into account task priority, estimated processing time, customer special requirements and market demand urgency factors For example, for order tasks with special surface roughness requirements and tight delivery time, higher priority weights are given, and task allocation plans are formulated in combination with machine tool maintenance plans and production capacity balance, high-precision tasks are assigned to machine tools with excellent performance, and task start times are arranged reasonably.

[0023] Using the biologically inspired intelligent scheduling algorithm, in the simulation of the pheromone communication mechanism of the ant colony algorithm, the pheromone concentration update formula (set up , , introducing a random perturbation factor (set up ) Avoid local optimality and penalize paths with low pheromone concentration (set , build a resource allocation model by imitating the division of labor and cooperation mode of honey bee colonies , refine the CNC machine tool resource attribute parameters, determine the weight coefficient based on expert experience and data analysis, generate accurate scheduling plans and send instructions to the machine tool for execution.

[0024] Embodiment 2 An automobile parts manufacturing company carries out large-scale standardized automobile engine parts production. The production volume is large and the process is relatively fixed. The product quality requirements are strict, and it is necessary to ensure that the production line operates efficiently and stably.

[0025] Highly reliable sensors are installed at key positions of CNC machine tools on the production line to collect spindle speed n and feed speed v at a frequency of 5 times per second (f1=5 times / second), tool wear w and machine load power P at a frequency of 2 times per minute (f2=2 times / minute), and production task information such as task number Ti, task type Ttype (uniformly used as engine parts processing type), processing technology requirements Tprocess (in accordance with automotive industry standard processes), estimated processing time Ttime (determined based on batch production experience), task priority Tpriority (set according to order delivery sequence), as well as raw material type Mtype (specific steel), inventory quantity Mnum, quality inspection parameter Mquality (strictly inspected in accordance with automotive parts quality standards). The data is verified, corrected and encrypted before being transmitted to the cloud computing processing layer.

[0026] A data cleaning strategy is adopted to remove outliers in numerical data, and character data is formatted and stored in a classified manner after standardization. The machine tool performance evaluation model E=α1n+α2v+α3w+α4P (α1=0.25, α2=0.25, α3=0.25, α4=0.25, optimized and determined based on a large amount of historical data) is used to monitor machine tool performance. At the same time, the machine tool fault prediction model is used to detect potential fault risks in advance. Through in-depth analysis of historical production data, the task planning layer is provided with machine tool operation status trend and raw material consumption forecast information to assist in formulating reasonable production plans.

[0027] When constructing the task priority sorting model O=∑i=1mλi×Fi, the main considerations are the estimated processing time Ttime and the task priority Tpriority (the priority is determined according to the order of order delivery time), and the weight coefficient λi is set to 0.6 and 0.4 respectively (determined by analyzing previous large-scale production orders). Tasks are allocated in a balanced manner in combination with machine tool maintenance plans and production capacity. For example, for machine tools that are about to undergo regular maintenance, their task allocation amount is reduced, and production tasks are reasonably allocated to machine tool groups with different performance and production capacity to ensure that the workload of each machine tool is balanced and the overall production efficiency is improved. At the same time, the task start time is arranged according to the raw material supply plan and inventory status to avoid shortages or backlogs of raw materials.

[0028] Intelligent scheduling execution layer: executes the intelligent scheduling algorithm. In the simulated ant colony algorithm, the pheromone volatility coefficient ρ=0.1 (determined according to the stable operation characteristics of the production line), the pheromone update formula is τij(t+1)=(1-ρ)τij(t)+Δτij(t), and the random disturbance factor ϵ (value range (-0.05, 0.05)) is introduced to prevent falling into the local optimum. The probability of selecting the path with low pheromone concentration is reduced to 0.8 times the original (k=0.8) to accelerate convergence. In the resource allocation model Ars=γ1Rr+γ2Ss+γ3Drs, for the CNC machine tool resource attribute parameter Rr, the factors of accuracy, power, maintenance record (focusing on the execution of regular maintenance) and service life are comprehensively considered to determine the weight coefficients γ1=0.35, γ2=0.35, and γ3=0.3 (combined with data analysis and industry expert experience), and generate scheduling instructions to control the efficient operation of CNC machine tools to ensure the quality and progress of large-scale standardized production.

[0029] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. The intelligent processing system of CNC machine tool automated production line based on cloud computing is characterized by: The system includes the following components: data acquisition layer, cloud computing processing layer, task planning layer and intelligent scheduling execution layer; The data acquisition layer is deployed at each key node of the CNC machine tool automation production line and is composed of a variety of sensors and data acquisition equipment. This layer is responsible for collecting all-round and real-time production-related data covering CNC machine tool operation data, production task information, and multi-dimensional raw material status. For CNC machine tool operation data, the collected parameters include but are not limited to the spindle speed. , Feed speed , tool wear , Machine tool load power , by installing corresponding sensors at key parts of the machine tool, obtaining production task information, and recording the task number in detail , Task Type , Processing technology requirements , Estimated processing time , Task Priority , raw material status data involves the type of raw materials , Inventory quantity , quality inspection parameters ,The collected data is transmitted in real time to the cloud computing processing layer in encrypted form through a high-speed and stable network transmission protocol; The cloud computing processing layer: relying on a powerful cloud computing platform, has large-scale data storage and high-performance computing capabilities. After receiving the data transmitted by the data acquisition layer, it first performs data cleaning operations on it, removes noise data and obviously erroneous data records by setting data threshold ranges and abnormal data recognition algorithms, and then classifies and organizes them according to data types and features, and stores the processed data in a distributed database. For different types of data, it uses specific data analysis algorithms for in-depth mining. For machine tool operation data, it establishes a machine tool performance evaluation model ,in , , , is the weight coefficient. According to the historical operation data of the machine tool and the actual processing experience, the hierarchical analysis method is used to determine the weight coefficients to evaluate the current performance status of the machine tool. For the production task information, the task complexity evaluation formula is used , As the weight coefficient, various types of information obtained from the analysis are integrated to provide comprehensive and accurate data support for the task planning layer; The task planning layer: Based on the detailed data support provided by the cloud computing processing layer, it conducts comprehensive analysis and planning of production tasks, and , Estimated processing time As well as the logical relationship between tasks, build a task priority sorting model ,in The comprehensive priority ranking result of the tasks is is the number of factors that affect task priority, is the weight coefficient of each factor, For the Factors that affect task priority. For urgent order tasks, priority factors are given higher weights. According to the performance evaluation results of the machine tool Based on the availability of resources, a preliminary task allocation plan is formulated to reasonably allocate tasks to different processing stages and machine tool groups, providing a task planning framework for the intelligent scheduling execution layer; The intelligent scheduling execution layer: based on the task allocation framework provided by the task planning layer and the analysis results of the cloud computing processing layer, uses the biologically inspired intelligent scheduling algorithm to generate an accurate scheduling plan. The algorithm combines the pheromone communication mechanism of the simulated ant colony algorithm and the division of labor and cooperation model of the bee colony. In terms of simulating the pheromone communication mechanism of the ant colony algorithm, the pheromone concentration update formula is defined: ,in express Time Machine Tool To the task The pheromone concentration, is the pheromone volatility coefficient, with a value range of (0, 1). Its optimal value is determined through experiments and data analysis according to the dynamic changes of the production environment and the characteristics of the task. for Time Machine Tool Complete the task The amount of pheromone released after , is the total amount of pheromone released, For machine tools Complete the task The path length is used to simulate the division of labor and cooperation mode of the bee colony to build a resource allocation model: ,in Representation Resources Assign to task The adaptability, For resources The own attribute parameters, For the task The demand parameters, For resources With the task The matching parameters, , , is the weight coefficient. The intelligent scheduling execution layer sends detailed scheduling instructions to each CNC machine tool for execution based on the results generated by the above algorithm.

2. The intelligent processing system for CNC machine tool automated production line based on cloud computing according to claim 1 is characterized in that: The data acquisition device of the data acquisition layer uses different acquisition frequencies for different types of parameters when collecting the operation data of the CNC machine tool. and feed speed For parameters with high real-time requirements and frequent changes, set the acquisition frequency to times / second, accurately collects data through high-speed data acquisition cards, and measures tool wear and machine load power For parameters that change relatively slowly, the acquisition frequency is set to The data is collected times / minute using a timed sampling method. To ensure the accuracy and reliability of the data, the collected data is verified and corrected in real time. The data is encoded and decoded using a redundant check algorithm and an error detection code. When a data error is detected, re-sampling or repair operations are automatically performed. The data acquisition equipment also has an adaptive acquisition function that can automatically adjust the acquisition frequency and accuracy according to the operating status of the machine tool and the characteristics of the production task.

3. The intelligent processing system for CNC machine tool automated production line based on cloud computing according to claim 1 is characterized in that: When performing data cleaning, the cloud computing processing layer adopts different cleaning strategies for different types of data. For numerical data, by setting upper and lower threshold ranges, outliers that exceed the range are removed. For character data, regular expression matching and semantic analysis methods are used to identify and correct spelling errors and data in irregular formats. In the process of data classification and organization, the data is divided into multiple different categories according to its source, nature and purpose, and an independent data table structure is established for each category. In addition, the cloud computing processing layer also has data backup and recovery functions, regularly backs up the processed data in full, and records data change operations in real time. When data is lost or damaged, the data can be restored quickly and accurately.

4. The intelligent processing system for CNC machine tool automated production line based on cloud computing according to claim 1 is characterized in that: The cloud computing processing layer uses data analysis algorithms to analyze machine tool operation data, in addition to establishing a machine tool performance evaluation model. In addition, a machine tool fault prediction model was built. This model is based on a machine learning algorithm. By training a large amount of historical machine tool operation data and fault records, characteristic parameters related to machine tool faults are extracted, and a support vector machine algorithm is used to establish a fault prediction model. ,in is the extracted feature parameter, The mapping function obtained through training is used to improve the accuracy and reliability of fault prediction by continuously optimizing model parameters and adjusting the combination of feature parameters. When a possible machine tool failure is predicted, early warning information is sent to the intelligent scheduling execution layer in a timely manner.

5. The intelligent processing system for CNC machine tool automated production line based on cloud computing according to claim 1 is characterized in that: The task planning layer builds a task priority sorting model When considering factors that affect task priority Perform more detailed analysis, in addition to task priority , Estimated processing time In addition, we also consider the special requirements of customer orders and the urgency of market demand. We give higher priority to orders with special process requirements or tight delivery dates. At the same time, we combine market demand data to give priority to product tasks that are urgently needed in the market. At the same time, through market research, customer feedback and comprehensive analysis of historical order data, we continuously optimize the weight distribution to make the task priority sorting more in line with actual production needs and market orientation.

6. The intelligent processing system for CNC machine tool automated production line based on cloud computing according to claim 1 is characterized in that: When formulating the preliminary task allocation plan, the task planning layer fully considers the maintenance plan and production capacity balance of the machine tools. For the machine tools that are about to undergo maintenance, the task volume is minimized when allocating tasks to avoid production interruptions due to maintenance. At the same time, by analyzing the performance parameters and historical processing data of the machine tools, the production capacity of different machine tools is quantitatively evaluated, and tasks are reasonably allocated to different machine tools. For high-precision and high-speed machine tools, tasks with high precision requirements and high processing difficulty are allocated. For ordinary machine tools, conventional processing tasks are allocated. In addition, the task planning layer will arrange the start time of the task according to the inventory situation and supply plan of the raw materials.

7. The intelligent processing system for CNC machine tool automated production line based on cloud computing according to claim 1 is characterized in that: The intelligent scheduling execution layer introduces a random disturbance factor when simulating the pheromone communication mechanism of the ant colony algorithm to prevent the algorithm from falling into a local optimal solution. During the pheromone update process, the pheromone concentration is Perform random adjustment, the adjustment formula is ,in is the random disturbance factor, and its value range is , is the disturbance intensity coefficient. According to the complexity of the production task and the scale of the production line, its optimal value is determined through experimental tests. At the same time, in order to speed up the convergence of the algorithm, the path with lower pheromone concentration is penalized to reduce its probability of being selected. The specific implementation method is as follows: when the pheromone concentration of a certain path is Below the set threshold When the probability of selecting Reduced to the original times, , in this way, the algorithm is guided to search for the global optimal solution faster.

8. The intelligent processing system for CNC machine tool automated production line based on cloud computing according to claim 1 is characterized in that: The intelligent scheduling execution layer builds a resource allocation model by imitating the division of labor and cooperation mode of the honey bee colony. When the resource's own attribute parameters are Further refinement, for CNC machine tools, resource attribute parameters include not only basic parameters of accuracy and power, but also maintenance records and service life of the machine tools. Maintenance records are quantified by the number of maintenance times and maintenance time interval indicators, and service life is calculated based on the production date and current time of the machine tools. Incorporating these factors into resource attribute parameters can more comprehensively evaluate the actual processing capacity and reliability of the machine tools. , , In addition to using correlation analysis and optimization algorithms to solve the problem, expert experience weights are also introduced. Industry experts are organized to evaluate the importance of resources and tasks, assign different weights, and then merge them with the weights obtained from data analysis to obtain the final weight coefficient.

Images